*3.5. SENP-Dependent Processing Determines SUMO Localization*

We next investigated the impact of SENP-dependent processing on the subcellular localization of SUMO. For this, promastigotes of six strains, *L. donovani* wild type, *L. donovani* SENP−/−, *L. donovani* (3×HA-SUMO), *L. donovani* SENP−/<sup>−</sup> (3×HA-SUMO), *L. donovani* (SUMO-3×HA) and *L. donovani* SENP−/<sup>−</sup> (SUMO-3×HA), from logarithmic culture, were spread on glass slides, fixed and stained with DAPI and with anti-HA tag antibody/anti-mouse AlexaFluor 594, followed by immune fluorescence microscopy. As expected, *L. donovani* wild type and *L. donovani* SENP−/<sup>−</sup> showed no 3×HA-specific staining (Figure 5A,D). We also did not observe a 3×HA-specific signal in *L. donovani* (SUMO-3×HA; Figure 5B), likely due to the cleavage of the C-terminal 3×HA tag in wild type cells. *L. donovani* (3×HA-SUMO) cells showed overlapping staining by DAPI and anti-HA tag antibody, indicating a nuclear localization of 3×HA-SUMO in the wild type.

**Figure 5.** Subcellular localisation of HA-tagged SUMO. Wild type (WT) (**A**) or *L. donovani* SENP−/<sup>−</sup> (**D**) cells expressing SUMO-3×HA (**B**,**D**) or 3×HA-SUMO (**C**,**F**). Cells were visualized by differential interference contrast (DIC), DAPI staining of nucleus and kinetoplast, and mouse anti-HA antibody/anti-mouse AlexaFluor 594. DAPI and anti-HA images were merged with 50% transparence. Size markers (5 µm) are shown in the DIC images; nucleus (n) and kinetoplast (k) are pointed out in the top DAPI image.

*L. donovani* SENP−/<sup>−</sup> (SUMO-3×HA) and *L. donovani* SENP−/<sup>−</sup> (3×HA-SUMO) both showed cytoplasmic staining. We conclude from this that (i) SENP−/<sup>−</sup> mutants cannot cleave off the C-terminal 3×HA tag (Figure 5E) and (ii) SENP-mediated cleavage of the C terminus is essential for nuclear localization of 3×HA-SUMO. The lack of SENP therefore prevents C-terminal processing of SUMO, preventing SUMO from attaining or maintaining a nuclear localization.

#### *3.6. Growth Phenotypes of SENP Null Mutants*

Given its critical function in SUMOylation, we tested the impact of SENP on the growth of *L. donovani* at different temperatures. *L. donovani*, *L. donovani* (Cas9/T7-RNAP), SENP−/<sup>−</sup> clones 1 and 2 and the SENP−/−/+ add-back strain were seeded at low density, and growth was then monitored over 72 h. Cell densities at 72 h were normalized, with wild type *L. donovani* set at 100% growth. At optimal growth conditions, 25 ◦C and pH 7.0, both SENP−/<sup>−</sup> null mutants showed a 50% reduced proliferation compared with wild type and the Cas9-expressing strain. This growth phenotype was reversed by ectopic SENP expression (Figure 6A). At 37 ◦C, we recorded less, but still significant growth reduction due to the loss of SENP (Figure 6B). This may indicate that SENP function and/or SUMO conjugation is more important at the lower temperature associated with the insect stage. − − − − − −

**Figure 6.** In vitro growth of wild type and mutant *L. donovani*. Cells were seeded at 5 × 10<sup>5</sup> /mL and grown either at 25 ◦C/pH 7.0 (**A**) or at 37 ◦C/pH 7.0 (**B**) for 72 h. Final cell densities were normalized against wild type growth (100%). Bars show the median cell growth. *n* = 6 (3 biol. repeats, 2 techn. repeats each). \*\* = *p* < 0.01; \* = *p* < 0.05 (U-test, two-sided).

We also tested the intracellular survival of SENP null mutants in mouse bone marrow-derived macrophages and found no differences in parasite loads compared with wild type parasites (A.B. and C.B., unpublished observations), consistent with a primary role for SENP in the promastigote stage.
